Heat shock proteins (Hsps) are highly conserved molecules mediating protection against lethal damage following various stress stimuli in prokaryotic and eukaryotic cells. Also under physiological conditions they support folding of non-native or misfolded proteins and prevent aggregation during proliferation and cellular differentiation (Hartl and Hayer-Hartl, Science 295 (2002), 1852-1858). The best characterized group of chaperones belong to the Hsp70 family. Like other stress proteins, Hsp70s are most efficient if they operate in concert with co-factors as cellular chaperone machineries. Together with J domain co-chaperones (i.e. Hsp40), they support protein folding and assist translocation across membranes (Pilon and Schekman, Cell 97 (1999), 679-682). Heat shock proteins (HSP) are also inducible by physiological processes including cell differentiation and development (Lindquist and Craig, Annu. Rev. Genet. 22 (1988), 631). Intracellular HSP functions not only as molecular chaperones, they are involved in antigen processing and presentation as well (DeNagel and Pierce, Immunol. Today 13 (1992), 86; Hartl et al., Nature 381 (1996), 571). HSP with a molecular weight of 70 and 90 kDa also have been shown to function as carrier proteins for immunogenic tumor-derived peptides that induce a T cell mediated immune response against cancer (Tamura et al., Science 278 (1997), 117; Schild et al., Current Opinion in Immunology 11 (1999), 109; Srivastava et al., Immunity 8 (1998), 657). Antigen presenting cells are key for the receptor mediated uptake of HSP-peptide complexes (Arnold-Schild et al., J. Immunol. 162 (1999), 3757). Several groups reported an unusual plasma membrane localization of HSP on tumor cells (Altmeyer et al., Int. J. Cancer 69 (1996), 340; Ferrarini et al., Int. J. Cancer 51 (1992), 613; Piselli et al., J. Biol. Regul. Homeost Agents 9 (1995), 55; Tamura et al., J. Immunol.
151 (1993), 5516). The inventors were the first who demonstrated that NK cells also have to be considered as relevant effector cells for the recognition of membrane-bound Hsp70 on tumor cells (Multhoff et al., Blood 86 (1995a), 1374; Multhoff et al., Int. J. Cancer 61 (1995b), 272; Multhoff et al., J. Immunol. 158 (1997), 4341; Botzler et al., Cancer Immunol. Immunother. 43 (1996a), 226; Botzler et al., Int. J. Cancer 65 (1996b), 633). With respect to these findings and due to the fact that normal cells lack expression of Hsp70, on the plasma membrane, one might speculate that Hsp70 acts as a tumor-selective recognition structure for NK cells. Antibody blocking studies revealed that Hsp70 is a relevant recognition structure for transiently plastic adherent NK cells (Multhoff et al., Blood 86 (1995a), 1374; Multhoff et al., Int. J. Cancer 61 (1995b), 272; Multhoff et al., J. Immunol. 158 (1997), 4341; Botzler et al., Cell Stress & Chaperones 3 (1998), 6).
It was recently demonstrated that proliferation and cytolytic activity of NK cells against Hsp70-expressing tumor cells could be stimulated with recombinant Hsp70 protein but not with Hsc70 or DnaK (Multhoff et al., Exp. Hematology 27 (1999), 1627). As target cells for the cytolytic activity of NK cells the tumor sublines CX+ and CX− with an identical MHC and adhesion molecule expression pattern that differ with respect to the capacity to express Hsp70 on the plasma membrane, were used (Multhoff et al., J. Immunol. 158 (1997), 4341).
As described above, the presence and localization of Hsp70 on the cell surface of diseased tissue or cells, in particular on tumor cells provides a valuable marker and target for therapeutic intervention. It is thus highly desirable to have antibodies or other binding molecules, which specifically recognize extracellular epitopes of Hsp70 on such tissue and cells.
Although several antibodies directed against Hsp70 are commercially available and have been described in the literature, these appear to be uncapable and/or unreliable in detecting membrane-bound Hsp70 on the surface of cells. The inventors tested a panel of these antibodies for the ability to detect plasma membrane bound Hsp70. Most of the tested antibodies were unsuitable for this task (SPA-820, Stressgen; H553220-clone7; BD Pharmingen; H5147 clone BRM-22; Sigma; 0A500 polyclonal, Dako; MS-482 clone W27, NeoMarkers), while others gave conflicting results.
The anti-Hsp70 antibodies from Affinity Bioreagents (MA3-006 and MA3-009) showed different specificities for different batches. Some batches were suitable for the detection of Hsp70 on the plasma membrane to some extent (Botzler et al., Cell Stress & Chaperones 3 (1998), 6) but recent batches showed no reactivity towards Hsp70 localized on the cell surface of tumor cells. Similarly, clone C92F3A-5 supplied by Stressgen Inc. as SPA-810 and by MBL, Japan, as SR-B810 has been described to react with cell surface localized Hsp70 on some occasions (Barreto et al., Cell. Immunol. 222 (2003), 97-104; Feng et al., Blood 100 (2002), 4108-4115) but the inventors were unable to repeat the reported results. In the report of Feng et al. cell surface Hsp70 was only detected in apoptotic cells, which may be due to antibodies entering the degrading apoptotic cells whose cell membrane was no longer intact and detecting intracellular Hsp70. The conflicting results of Baretto et al and other investigators might further be due to quality differences of the batches used for the respective experiments. Also there are no reports of binding cell surface Hsp70 so far for the MBL supplied antibody.
Loss of specificity or a change of specificity of a monoclonal antibody may also be due the hybridoma cell line producing said antibody not being derived from a single cell line. A mixture of two or more different hybridomas will produce a mixture of two or more monoclonal antibodies with different specificity. The ratio of the cells within the culture and thereby the ratio of the different monoclonal antibodies produced by them may vary during cultivation. Also, a particular hybridoma producing the antibody with the desired specificity might be lost from the mixture if the other hybridoma cells have an evolutionary advantage. A mixture of hybridomas in the culture can also result from mutations in certain cells leading to shifting specificities of the antibodies produced.
Thus, there is a need for a reliable source of anti-Hsp70 antibodies capable of detecting extracellular epitopes of Hsp70 and thereby enabling the specific detection and treatment of tumor cells or cells infected by a pathogen.
The solution to said technical problem is achieved by providing the embodiments characterized in the claims and described further below.